Journal Logo


Effects of Ibuprofen during Exertional Heat Stroke in Mice


Author Information
Medicine & Science in Sports & Exercise: September 2020 - Volume 52 - Issue 9 - p 1870-1878
doi: 10.1249/MSS.0000000000002329


There is widespread use of nonsteroidal anti-inflammatory drugs (NSAID) throughout the United States. For example, by 2004, approximately 8% to 12% of the entire US population older than 40 yr was regularly taking nonaspirin NSAID (1). In the US Army, by 2014, 82% of active duty Army received prescriptions for NSAID (2), with ibuprofen (IBU) comprising approximately 49% of all drug prescriptions (2). In one sample of collegiate football, approximately 70% of athletes were regularly taking NSAID during the season, and of these, approximately 80% were using IBU (3).

There are concerns with heavy use of nonsteroidal anti-inflammatory drugs in individuals exercising intensely in the heat. Ibuprofen is considered a nonspecific NSAID, suppressing both cyclooxygenase 1 (COX-1) and COX-2 (4), but with approximately 15 times greater potency against COX-1 (5). This results in notable effects on the gastrointestinal (GI) barrier, reviewed in Bjarnason et al (6). In addition, IBU has other important side effects. For example, IBU ingestion in long endurance events is associated with higher incidences of kidney injury (7) and the US Food and Drug Administration recently issued new warnings regarding increased risks of thrombotic cardiovascular disease and stroke with frequent NSAID use, guidelines that include warnings regarding IBU (8).

Both intense exercise and hyperthermia may exacerbate the effects of NSAID on the GI tract. Acute GI disturbances during competition are frequently experienced among endurance athletes, which is often associated with GI barrier dysfunction (9). Barrier dysfunction is aggravated proportionately in humans with increasing core temperature (Tc) (reviewed in [10]) and by preingestion of NSAID, such as IBU (11,12). Hyperthermia alone can induce GI injury in animal models of passive heat stroke (13,14) and exertional heat stroke (EHS) (15). When indomethacin, a predominant COX-1 inhibitor, was given at a high dose immediately before exposure to passive heat stroke in mice, 47% of the animals died compared with no deaths in controls or compared with mice given very low dose indomethacin (16). The mice treated with high-dose indomethacin showed clear evidence of overt GI bleeding. Therefore, concern over NSAID use before exercising in the heat appears warranted, but the specific effects of different NSAID (particularly IBU) in conditions of heat stroke are not well defined.

Because IBU is by far the NSAID of choice in the United States for both athletes and active military personnel, whom often exercise in hot environments, we tested whether IBU treatment affects overall exercise performance and whether it contributes to further development of GI barrier dysfunction in an animal model of EHS. Important differences from previous work include: 1) the use of an EHS model in mice with neurological endpoints resembling clinical definitions of heat stroke in humans (15). The endurance exercise component of the model may have unique influences on GI barrier function compared with models of passive heat stroke. 2) The incorporation of IBU into the chow of the animals for a 48-h period before the EHS exposure. This method ensured that the IBU, in principle, could be distributed throughout the GI tract, that it was not associated with GI injury or stress from gavage procedures, as well as being mixed with a normal fiber content. The latter is a critically important variable in NSAID-induced GI injury (17). This administration method more closely resembles the prescribed administration for humans in the United Kingdom (18) and precautionary notes on the packaging label in the United States (19), that is, “taken with food.” 3) The use of a dosage of IBU within chow that has previously been identified in male C57bl/6 mice to be tolerated well and to be highly effective in reducing pain and platelet thromboxane production in serum samples (20). 4) Comparisons of treatment between both males and females undergoing EHS, as male and female mice have profound differences in exercise performance leading up to EHS (21), and sex differences have been identified regarding sensitivity to IBU (22). Our working hypothesis was that IBU treatment would reduce exercise performance in the heat in both sexes and exacerbate the extent of GI damage that occurs in this model.


Animal subjects

This study was approved by the University of Florida’s Institutional Animal Care and Use Committee and by the Animal Care and Use Committee of the US Army Medical Research and Material Command. All mice were C57BL/6J (Jackson Laboratories, Bar Harbor, ME) and upon arrival, housed on a 12:12-h light/dark cycle at 20°C to 22°C and 30% to 60% relative humidity (RH). Standard chow (2918 Envigo; Teklad, Madison, WI) and water were provided ad libitum. All mice were approximately 12 to 13 wk old upon arrival and were approximately 18 to 22 wk at the end of the protocol. Mice were randomized into four groups for the EHS experiments: EHS only, EHS + IBU, exercise only (EXC), EXC + IBU. EXC mice were match-controlled to EHS mice in terms of the duration of the exercise protocol and time of tissue collection, except they were kept at 22.5°C and approximately 50% RH during exercise.

Surgery and exercise training

Implantation surgery of telemetry devices was performed under sterile conditions as previously described (15). Under isoflurane anesthesia, a laparotomy was performed to place temperature telemetry emitters into the abdominal cavity (G2 E-Mitter; Starr Life Sciences, Oakmont, PA) for measurements of Tc. The abdominal cavity was closed, and mice were single-housed during recovery and for the rest of the experiment. They were monitored for pain and behavior for 48 h after surgery, and subcutaneous injections of buprenorphine were given every 12 h for analgesia. Mice recovered from surgery for 2 wk and then were given modified in-cage running wheels (model 0297–0521; Columbus Instruments, Columbus, OH) to allow for voluntary exercise training for 3 wk. During the third week, animals were exercised on a forced running wheel on four occasions over four consecutive days (Lafayette, model 80840, Lafayette, IN; powered by a programmable power supply, as described previously [21]). Once training was completed, mice were given two to three full days of rest before either the EHS or EXC trial. In-cage running wheels were made available during that time.

Feeding protocol

Ibuprofen was administered within the feed as described by Salama et al. (20). Food was weighed to confirm the feeding rate for each mouse. The feed was prepared in 2918 Envigo chow with USP IBU using a commercial enrichment company (Bio-Serv, Flemington, NJ). Mice given IBU began dosing with IBU 48 h before the scheduled EHS protocol. The mice given IBU in chow consumed 4.4 ± 0.8 (males) and 4.6 ± 0.7 g·d−1 (females), values nearly identical to published work on expected daily food intake in C57bl/6 mice of this age (23,24). The dose used was 375 mg·kg−1, the highest dose previously tested to be effective by this route (20). We removed the original food from EHS-only mice and replaced it with identical feed but without IBU (placebo).

EHS protocol

The EHS protocol was carried out as shown in the idealized schematic in Figure 1. Mice were brought to the laboratory the afternoon before EHS. Tc was recorded in 30-s intervals throughout the night to ensure normal temperature profiles before EHS. The 12:12-h light/dark cycle was maintained. The EHS procedure was run in the early morning beginning at approximately 7:30 am. Mice remained in their cages with Tc being monitored while the environmental chamber (Thermo Forma, 3940; Thermo-Fisher, Waltham, MA) heated to a set point of 37.5°C, and 30% to 40% RH. Environmental chamber temperature and humidity were measured and recorded at the location of the running wheel. Once the temperature equilibrated (30–45 min), the mice were placed in the enclosed forced running wheel within the chamber. Mice were given >5 min to recover from the stress of being handled and then, once Tc stabilized again to 36°C to 37.5°C, the running wheel was started on a preprogrammed and standardized incremental protocol (Fig. 1). The average initial Tc of the mice before running were not different between groups, as reported in Supplemental Table 1 (see Table, Supplemental Digital Content 1, Initial Tc and Body weights, Speed began at 3.1 m·min−1 and increased 0.3 m·min−1 every 10 min until the mouse reached a Tc of 41°C. Once 41°C Tc was reached, the “steady-state” exercise phase began with speed maintained until the symptom-limited end point (Fig. 2). The EHS end point for this model was previously defined by loss of consciousness (15), specifically, three consecutive revolutions of the wheel with no physical response by the mouse (21). After symptom limitation, mice were checked for responsiveness to tactile touch and then quickly removed from the exercise wheel and placed in their cage with ad libitum access to food and water. The environmental chamber door was opened to room air during the recovery period and the temperature regulator for the incubator changed to room temperature. Tc was recorded continuously until tissue collection at 3 h post-EHS.

Experimental design of EHS exposure and time of tissue collection. The first component of the treatment was an incremental exercise stage that lasted until the mouse attained a Tc of 41°C. The mouse then performed steady-state exercise until it reached symptom limitation. The duration from the beginning of 41°C until reaching Tc,max and symptom limitation was considered the ability of the mouse to resist heat stroke. Samples were collected after 3 h of recovey.
Effects of IBU treatment on A: the time of steady state exercise during hyperthermia >41°C, and indicator of resistance to heat stroke (Wilcoxin) and B: peak Tc attained during the EHS (two-sample t test) and C: The total exercise time in the heat before symptom limitation (two-sample t test) (n = 14–16 EHS + IBU, n = 6 EHS only); means ± SD.

Sample collection and analyses

At 3 h post-EHS, while under isoflurane anesthesia, whole blood samples were drawn via transthoracic cardiac stick, using a 27-gauge needle, preloaded with ethylenediaminetetraacetic acid. Animals were then euthanized under deep isoflurane anesthesia by excision of the heart. Hematology analysis was immediately done on whole blood using a HESKA Element HT-5 Veterinary Hematology Analyzer according to manufacturer’s instructions (HESKA Corp., Loveland, CO). Whole blood was then centrifuged at 4°C for 10 min at 2000g, the plasma transferred to microtubes and immediately snap frozen in liquid nitrogen. Gastrointestinal tissues were immediately obtained for histological analyses. The 3-h timepoint was used because of previous experience with development of GI injury in this model (15). The entire surface of the stomach was evaluated for microhemorrhage or other abnormalities. The small intestines were prepared using a bundling method as described by Williams et al. (25). Sections of the duodenum, jejunum, and ileum were independently analyzed. Histological slides were prepared by the University of Florida Molecular Pathology Core Laboratory (4 μm thick). Imaging was done with a 10× objective and when the entire intestine was too big for one image field, multiple images were merged together for analyses. The original Chiu method was used for scoring villi size and injury (26). Measurements of damage, intestinal villi, and crypt depths were obtained and averaged from two blinded graders. Surface area was calculated from these measurements as (2π) (VW / 2) (VL), where VW is villi width and VL is villi height (27). This was the primary outcome variable reported. Gastrointestinal damage was further verified using a biomarker for small intestinal damage, intestinal fatty acid binding protein-2 (FABP2). An ELISA kit for PGE2 (R&D Systems, Minneapolis, MN) and FABP2 (Cloud-clone Corp., Houston, TX) were used with plasma samples following manufacturer’s instructions. The minimum detectable dose for the FAPB2 ELISA was 0.059 ng·mL−1. The intra-assay coefficient of variation (%CV) was <10% and interassay %CV was <12%. For PGE2, the minimum detectable dose was 31 pg·mL−1; the intra-assay CV was <10% and the interassay CV was <13%.

Statistical and analytical approach

G*Power was used to calculate a desired sample size. The “effect size” was determined using the variance of max Tc as the variable of interest, resulting in an n = 6 as the minimum sample size. To overcome expected sampling errors (e.g., difficulties in blood sampling, errors in intestinal histology preparation, temperature transmitter errors, etc) an initial n = 8 was targeted for most groups. However, in the original design, there were two different sets of time-matched groups, where two groups of animals were run simultaneously against each other with identical exercise protocols, that is, (IBU + EHS vs IBU + EXC) and (IBU + EHS vs EHS only). After analysis was complete, there was no statistical advantage of pairing these measurements; therefore we simplified the outcomes by combining all IBU-EHS–treated mice into single groups in each sex. This resulted in a nonsymmetrical group sizes, ranging from 14 to 16 for IBU + EHS and 6 to 8 for all other groups.

Statistical testing and graphics were performed using SAS-JMP ® and/or GraphPad Prism. The Shapiro–Wilk test was used to examine normality of populations and Fisher test for equality of variances. For single measurements of performance in IBU-EHS versus EHS animals, data were analyzed using a two-sample t test for parametric data with equal variances or Welch’s correction for unmatched variances. For multiple groups of parametric data, one- or two-way ANOVA and appropriate post hoc analysis with the Welch’s correction used for unequal variances. For nonparametric samples, the Wilcoxon signed ranks test was used. In all cases, details of statistical tests used are stated in the figure legends.


Effects of IBU on exercise performance in hyperthermia

As reported previously (21), females consistently outperformed the males in terms of their running endurance (Fig. 2C) and speed attained during the EHS trial (not shown). However, as shown in Figure 2A, male mice treated with IBU exhibited approximately 87% longer exercise time during the last steady-state phase (Tc >41°C) compared to placebo-treated male EHS mice. Male mice treated with IBU also reached symptom limitation at a higher Tc,max compared with untreated EHS controls (mean = 42.31 vs 41.97, P < 0.01), Figure 2B). In females, effects of IBU treatment did not reach statistical significance for either time > 41°C or Tc,max. Of cautionary note: overall, males reached a somewhat lower Tc,max compared with females. There were no effects of IBU treatment on initial body weight or the percentage of body weight lost during EHS (see Table, Supplemental Digital Content 1, Initial Tc and Body weights, Ibuprofen also had no impact on the degree or timing of hypothermia during the 3-h recovery from EHS (data not shown).

Effects of IBU on GI damage

Figure 3 illustrates the effect of EHS or EXC on a plasma biomarker for intestinal epithelial injury, FABP2. Exertional heat stroke elevated FABP2 in both males and females compared with exercise-matched controls, but there were no additional effects of IBU treatment in either group. Females, in general, had higher levels of FABP2 during EHS compared with males (P < 0.02).

Effects of IBU and EHS on plasma fatty acid binding protein, (FABP2), a biomarker for intestinal injury. P = placebo, Exer = time-matched exercise-matched control mice. Means ± SD, Welch’s one-way ANOVA corrected for unequal variances, post hoc test comparisons were Dunnett’s for multiple comparisons (n = 14–16 EHS + IBU, n = 6 EHS only; means ± SD).

Histological measurements of GI damage were evaluated using average villi surface area. In exercise controls, IBU treatment in both males and females consistently resulted in smaller villi throughout the three regions of the small intestine (Fig. 4A and B), which is an indicator of recent damage and rapid restitution. After EHS exposure, males demonstrated further reductions of villi size in the duodenum, as expected (15). Importantly, IBU treatment in EHS animals had no significant additional effects on villus size in any region, in either sex. Measurements of gastric injury (microbleeding) and intestinal villi damage scores were not significantly different in response to IBU in any group (data not shown).

Effects of IBU treatment in A and B: exercise controls and C and D: EHS animals. Measurements of average intestinal villi surface area in each region of the small intestine (n = 6 mice per group, randomly selected for EHS + IBU). Results of two-way ANOVA are inserted into each panel. Individual P values are post hoc contrasts.

Effects of EHS and IBU on immune cell populations

Acute exercise (EXC) resulted in significant reductions in total white blood cell (WBC) counts compared with identically treated control animals allowed to recover for 4 d (Fig. 5A, males). Samples for female 3 h exercise controls were not available due to equipment failure. There were no significant effects of acute exercise on %neutrophils, %lymphocytes, or %monocytes (Fig. 5B, C, D, males). However, EHS exposure reduced total WBC counts, greatly elevated the %neutrophils in both males and females (Fig. 5B), reduced the %lymphocytes and elevated % monocytes.

Effects of ibuprofen (IBU), and or exertional heat stroke (EHS) in whole blood on (A) whole blood cell counts, (B) %Neutrophils, (C) %Lymphoctyes and (D) %Monocytes. Note the addition of a 3 h EX control group is provided for females only because of technical difficulties. Results for each sex are from ANOVA with Welch’s correction for unequal variances, followed by Tukey’s (n = 14–16 EHS + IBU, n = 6 EHS only or 3 h EXC; means ± SD).

The IBU treatment in exercise control females significantly elevated the total blood WBC counts (Fig. 5A). However, there were no other significant effects of IBU on % cell phenotypes in EXC or in EHS.

Effects of IBU treatment on serum PGE2

Plasma PGE2 was measured at the time of sample collection (3 h post-EHS) to test the efficacy of the IBU at this timepoint (Fig. 6A). In exercise controls (Fig. 6A), IBU-treated male mice showed marked inhibition of plasma PGE2 (144 ± 126 ng·mL−1 in EXC vs 33 ± 28 in EXC + IBU). We did not observe such inhibition in females (Fig. 6B) but females had a much higher overall PGE2 in plasma. In males, EHS independently suppressed plasma PGE2, with IBU treatment having no additional effects. In females, there was no effect of either EHS or IBU on plasma PGE2.

Plasma [PGE2] in A: females and B: males in placebo (P) controls vs IBU-treated mice. The sample size was smaller for this test because of lack of available plasma. Data were skewed and therefore transformed before statistical testing (n = 10 EHS + IBU, n = 6 for all other groups; means ± SD). 1-Way ANOVA for each sex, followed by Sidak’s multiple comparison test.


In contrast to our working hypothesis, we found that IBU treatment did not reduce exercise performance in the heat. In fact, male mice significantly increased their tolerance to acquiring heat stroke. This manifested as a higher attained Tc,max at symptom limitation and an increased duration of exercise during the last steady-state stage (>41°C). In contrast, in female mice, there was no statistical effect of IBU on performance. The IBU treatment in control mice of both sexes exhibited the expected injury throughout the small intestine. However, after EHS there was no additional GI injury induced by IBU treatment in either sex. This result suggests that the IBU administration protocol was effective in delivering the drug to the target, but the IBU stimulus was either not strong enough to induce further damage or some aspect of exercising in the heat prevented IBU from further contributing to damaged villi.

IBU administration

We faced several challenges in using an effective drug delivery method in this experiment. We considered using intraperitoneal injection, oral gavage, mixed in drinking water, mixed in chow and mixed in a single dose within a treat. Because our interest was in studying effects on the GI tract, the intraperitoneal route was not an option. In our experience, oral gavage is extremely stressful to mice and can delay gastric emptying. Therefore, providing the IBU with food, over time (20), would more closely mimic conditions experienced by athletes and the military when they take it on a daily basis.

This tactic avoided peak serum levels, allowed better distribution through the GI tract, and included the presence of dietary fiber, which can both contribute to and suppress small intestinal damage induced by NSAID, depending on the type of fiber, and the effective bioavailability of the drug (28). Furthermore, the dose given was previously shown to be the high end of an effective delivery method to reduce pain over several hours after ingestion (20).

Timing and rate of metabolism of IBU was also a concern. Mice eat primarily during their waking hours, that is, intermittently throughout the night. Our mice are on a 7:00/7:00 light cycle, and we began their EHS protocol or exercise protocol at approximately 7:30 am with a 30 min equilibration phase. The entire EHS protocol then requires an average of 2 h in males and 3 h in females (21), with a 3-h recovery period. Therefore, IBU had to be active for a minimum of 3 to 4 h to impact the mice during the EHS exposure. The half-life of IBU in the blood of mice when given orally in water is ≈170 min (29), but the ability to inhibit responsiveness to pain continues for at least 6 h (30). The fact that we observed clear reductions in PGE2 (Fig. 6) in male exercise control mice when measured in the blood at approximately 7 h after their last ingestion confirms that the drug was likely to be effectively working at this time. However, we are less confident that it was still effective in females, where tissue sample collection was yet another 1.5 h beyond the males because of the much greater exercise capacity and resistance to heat stroke (21).

The relevance of providing IBU in the food with respect to how it is self-administered by athletes or military personnel is also a concern. It is likely that IBU is often taken on an empty stomach and at high dosages, where the effects are likely to be more profound. For example, in a cross section of different sports, 12% to 37% of collegiate athletes report taking more than the recommended dose (3,31). Much less information is available regarding how the drugs are taken relative to food intake, but it is apparently common for IBU to be taken on an empty stomach before, and even during athletic events (32,33). The presence of food in the GI tract can have considerable impact, e.g. in humans, the absorption of IBU in the fasting state is approximately 1.4 h versus 1.6 h in the fed state, with peak plasma concentrations reaching approximately 28% higher in the fasted state (34). Perhaps, more importantly, the presence of food directly influences the integrity and crosslinking of the mucosal layer and thus provides additional protection from the negative effects of NSAID (28,35).

A number of studies have demonstrated that IBU will increase intestinal permeability during athletic activities in humans (11,12) where IBU was apparently taken in the fasting state. Our results, in essence, are in agreement with these findings, in that IBU taken in food caused marked small intestinal damage in the EXC mice in the absence of hyperthermia. The notable contribution of this study was that the IBU, given before EHS, did not create further damage over and above the considerable GI damage caused by heat exposure.

Effects of IBU on performance

The males treated with IBU were better able to resist the onset of heat stroke compared to their placebo-treated controls (Fig. 2). We have no direct evidence for a mechanism but wish to discuss several possibilities. First, there are mixed reports that IBU and other related analgesics have the capacity to improve endurance exercise performance in both rodents and humans. In exercise-trained rats, IBU increased time to exhaustion using a swimming model (36). Another study showed that chronic IBU treatment results in longer distances of voluntary wheel running in mice (37). However, in human studies there is sparse evidence that IBU improves endurance exercise (38–40). In contrast, several human studies have shown that acetaminophen administration results in greater performance during endurance exercise (41,42). Most notably, this effect is seen during cycling exercise to exhaustion in a hot environment (30°C/50% RH), where endurance time was increased and skin temperature and Tc reduced by acetaminophen (43). Though not technically categorized as a NSAID, acetaminophen is a well-known PGE2 inhibitor and an antipyrogenic. One potential mechanism that has been proposed is that acetaminophen and IBU can cross the blood brain barrier and reduce brain PGE2 levels. PGE2 acts as a pyrogen in the brain, elevating the thermoregulatory set point, particularly in conditions of fever. Some investigators have proposed that pyrogenic effects of brain eicosanoids contribute to elevations in body temperature during endurance exercise (43).

A further consideration that relates specifically to this model is that the males exercised to a higher Tc compared with their placebo-treated counterparts (Fig. 2), which was unexpected. Based on the pyrogen hypothesis discussed above, this would appear to be an opposite effect. However, it is important to note that the endpoint in this model is CNS dysfunction, not exercise exhaustion. We do not understand the origins of CNS dysfunction in heat stroke but it could reflect loss of cardiovascular function (44), attaining a pain or stress threshold, loss of blood–brain barrier integrity or low nutrient delivery, all of which could be influenced in various ways by NSAID and analgesics.

Low responsiveness in females

In females, we saw no statistically significant effect of IBU treatment on exercise performance, no effect on PGE2 levels at the time of sample collection, and no apparent effects on immune cell profiles at any time. However, IBU-induced injury to the small intestine in the time matched exercise controls was nearly identical in both sexes (Fig. 4A and B), indicating that the drug delivery system was effective in both. These results suggest that either the longer exercise time experienced by the females (1.5–2 h) reduced the bioavailability of IBU during the late stages of EHS or that female mice had a lower sensitivity to IBU. There is no available evidence for lower sensitivity to IBU in female rodents, but differences in effectiveness of IBU in pain management have been observed in females (22). Nevertheless, we believe the most likely explanation for our different results between the males and females is that the exercise time of the females extended beyond a period of IBU effectiveness.

Another interesting sex difference was that in male mice, post-EHS, there was a marked suppression of plasma PGE2 in IBU-EXC mice and in all EHS mice (Fig. 6A), but the effect was not seen at all in females (Fig. 6B). Similar reductions in PGE2 have been observed after human EHS in military populations, which were mostly male (45). The effects of heat and stress on eicosanoid production are complex, but there are several hormonal influences that may be relevant. For example, we have previously shown that both male and female mice respond to EHS with 400% to 500% increases in plasma corticosterone before 3 h of recovery (21). Glucocorticoids have been shown to inhibit PGE2 production by several known mechanisms (46). Nevertheless, the high PGE2 levels in females, that appeared unaffected by either EHS exposure or IBU treatment (Fig. 6), remain a mystery. Though plasma PGE2 is known to be similar in control male and female mice, COX-2 is under regulation by estradiol (47) which can result in sex differences. For example, PGE2 elevations after the stress of burn injury in female mice are much higher compared with males (48), an effect attributed to a reduced rate of PGE2 degradation and clearance in stress due to reductions in estradiol-regulated prostaglandin dehydrogenase.

Lack of effects IBU on small intestine injury after EHS

Another unexpected finding was that although IBU caused injury to the small intestine in all exercise control mice, there was no evidence of further injury in IBU treated EHS mice. It could be that we met some threshold for injury from EHS that was unaffected by superimposed IBU. However, another possibility is that the normal hormonal responses to EHS as compared with control exercise conditions may have prevented further loss of gut integrity. In a previous study we reported a rapid spike in IL-6 production immediately after EHS (21). We have also previously shown that the preconditioning injections of IL-6 protect the intestinal lining during the recovery period after EHS (14). Also, elevations in endogenous glucocorticoids at this stage (21) are protective against gastric injury induced from NSAID (49), though their effect on the small intestine is not as well defined. Finally, in the female mouse, suppression of PGE2 degradation during stress may have further protected the intestine and/or promoted its rapid recovery (50). Therefore, we suggest that a confluence of factors related to exercising in the heat induced an overriding protection of the intestine from further NSAID-induced injury in EHS or the effect of EHS on the gut is so severe that it masked any additional influence of IBU.


The results of these experiments are not consistent with our working hypothesis that IBU ingestion at analgesic doses before exercising in the heat necessarily comprises an added risk factor for developing heat stroke or suffering from further GI injury. However, these conclusions should be viewed with caution because humans may respond differently, human subjects often ingest higher than recommended doses of NSAID and ingest NSAID on an empty stomach. In addition, some individuals may have extenuating conditions such as underlying GI sensitivity that could make them vulnerable to negative effects of NSAID. The data in mice also does not support the use of NSAID in humans to suppress the onset of heat stroke or to improve exercise performance in the heat. Because of the many unknown factors in human use, we are a long way from fully understanding the positive and negative effects of NSAID ingestion in the competing athlete or the active Warfighter.

The authors would like to thank the University of Florida Molecular Core Pathology laboratory for their help in producing quality histological images and Lisa Leon from USARIEM for her help in designing the study and reviewing the article. This research was supported by a contract from the Department of Defense, BAA Extramural Medical Research W81XWH-15-2-0038 (T. L. C.) with supplemental support from the B. K. and Betty Stevens Endowment at the Univ. of Florida (T. L. C.).

The authors declare that the results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The authors have no conflicts of interest with any aspect of this manuscript. The results do not constitute endorsement by ACSM. The opinions or assertions contained herein also are the private views of the author(s) and are not to be construed as official or as reflecting the views of the Army or the Department of Defense. Citations of commercial organizations and trade names in this report do not constitute an official Department of the Army endorsement or approval of the products or services of these organizations.


1. Davis JS, Lee HY, Kim J, et al. Use of non-steroidal anti-inflammatory drugs in US adults: changes over time and by demographic. Open Heart. 2017;4(1):e000550.
2. Walker LA, Zambraski EJ, Williams RF. Widespread use of prescription nonsteroidal anti-inflammatory drugs among U.S. Army active duty soldiers. Mil Med. 2017;182(3):e1709–12.
3. Wolf DA, Miller TW, Pescatello LS, Barnes C. National Collegiate Athletic Association Division I Athletes’ use of nonprescription medication. Sports Health. 2011;3(1):25–8.
4. Bastaki SMA, Padol IT, Amir N, Hunt RH. Effect of aspirin and ibuprofen either alone or in combination on gastric mucosa and bleeding time and on serum prostaglandin E2 and thromboxane A2 levels in the anaesthetized rats in vivo. Mol Cell Biochem. 2018;438(1–2):25–34.
5. Mitchell JA, Akarasereenont P, Thiemermann C, Flower RJ, Vane JR. Selectivity of nonsteroidal antiinflammatory drugs as inhibitors of constitutive and inducible cyclooxygenase. Proc Natl Acad Sci U S A. 1993;90(24):11693–7.
6. Bjarnason I, Scarpignato C, Holmgren E, Olszewski M, Rainsford KD, Lanas A. Mechanisms of damage to the gastrointestinal tract from nonsteroidal anti-inflammatory drugs. Gastroenterology. 2018;154(3):500–14.
7. Lipman GS, Shea K, Christensen M, et al. Ibuprofen versus placebo effect on acute kidney injury in ultramarathons: a randomised controlled trial. Emerg Med J. 2017;34(10):637–42.
8. Safety Announcement: FDA strengthens warning that non-aspirin nonsteroidal anti-inflammatory drugs (NSAIDs) can cause heart attacks or strokes 2015 Available from:
9. van Wijck K, Lenaerts K, van Loon LJ, Peters WH, Buurman WA, Dejong CH. Exercise-induced splanchnic hypoperfusion results in gut dysfunction in healthy men. PLoS One. 2011;6(7):e22366.
10. Pires W, Veneroso CE, Wanner SP, et al. Association between exercise-induced hyperthermia and intestinal permeability: a systematic review. Sports Med. 2017;47(7):1389–403.
11. Lambert GP, Boylan M, Laventure JP, Bull A, Lanspa S. Effect of aspirin and ibuprofen on GI permeability during exercise. Int J Sports Med. 2007;28(9):722–6.
12. Van Wijck K, Lenaerts K, Van Bijnen AA, et al. Aggravation of exercise-induced intestinal injury by ibuprofen in athletes. Med Sci Sports Exerc. 2012;44(12):2257–62.
13. Lambert GP, Gisolfi CV, Berg DJ, Moseley PL, Oberley LW, Kregel KC. Selected contribution: hyperthermia-induced intestinal permeability and the role of oxidative and nitrosative stress. J Appl Physiol (1985). 2002;92:1750–61.
14. Phillips NA, Welc SS, Wallet SM, King MA, Clanton TL. Protection of intestinal injury during heat stroke in mice by interleukin-6 pretreatment. J Physiol. 2015;593(3):739–52; discussion 753.
15. King MA, Leon LR, Mustico DL, Haines JM, Clanton TL. Biomarkers of multiorgan injury in a preclinical model of exertional heat stroke. J Appl Physiol. 2015;118(10):1207–20.
16. Audet GN, Dineen SM, Stewart DA, et al. Pretreatment with indomethacin results in increased heat stroke severity during recovery in a rodent model of heat stroke. J Appl Physiol (1985). 2017;123(3):544–57.
17. Satoh H, Kondo R, Shinoda T, Idaka S, Ishigami K, Shiotani S. Diets with no or low amounts of dietary fiber can reduce small intestinal ulcers induced by non-steroidal anti-inflammatory drugs in dogs. J Physiol Pharmacol. 2016;67(4):563–73.
18. National Health Service United Kingdom. Ibuprofen for adults: painkiller. 2018; [cited 2019 Dec 26] Available from:
19. FDA Research Center for Drug Evaluation. Ibuprofen Drug Facts Label [Internet]. FDA. 2018; [cited 2019 Dec 26]; available from:
20. Salama RAM, El Gayar NH, Georgy SS, Hamza M. Equivalent intraperitoneal doses of ibuprofen supplemented in drinking water or in diet: a behavioral and biochemical assay using antinociceptive and thromboxane inhibitory dose-response curves in mice. PeerJ. 2016;4:e2239.
21. Garcia CK, Mattingly AJ, Robinson GP, et al. Sex-dependent responses to exertional heat stroke in mice. J Appl Physiol (1985). 2018;125(3):841–9.
22. Butcher BE, Carmody JJ. Sex differences in analgesic response to ibuprofen are influenced by expectancy: a randomized, crossover, balanced placebo-designed study. Eur J Pain. 2012;16(7):1005–13.
23. Bachmanov AA, Reed DR, Beauchamp GK, Tordoff MG. Food intake, water intake, and drinking spout side preference of 28 mouse strains. Behav Genet. 2002;32(6):435–43.
24. Starr ME, Saito H. Age-related increase in food spilling by laboratory mice may lead to significant overestimation of actual food consumption: implications for studies on dietary restriction, metabolism, and dose calculations. J Gerontol A Biol Sci Med Sci. 2012;67(10):1043–8.
25. Williams JM, Duckworth CA, Vowell K, Burkitt MD, Pritchard DM. Intestinal preparation techniques for histological analysis in the mouse. Curr Protoc Mouse Biol. 2016;6(2):148–68.
26. Oltean M, Olausson M. The Chiu/park scale for grading intestinal ischemia–reperfusion: if it ain’t broke don’t fix it! Intensive Care Med. 2010;36(6):1095–5.
27. Solis de los Santos F, Donoghue AM, Farnell MB, Huff GR, Huff WE, Donoghue DJ. Gastrointestinal maturation is accelerated in Turkey poults supplemented with a mannan-oligosaccharide yeast extract (Alphamune). Poult Sci. 2007;86(5):921–30.
28. Satoh H. Role of dietary fiber in formation and prevention of small intestinal ulcers induced by nonsteroidal anti-inflammatory drug. Curr Pharm Des. 2010;16(10):1209–13.
29. Komori K, Fukuda M, Matsuura T, et al. Effect of alcoholic beverages on drug absorption: blood concentration profile of ibuprofen in mice. J Appl Pharm. 2017;9(2).
30. Saad SS, Hamza M, Bahr MH, Masoud SI. Nitric oxide is involved in ibuprofen preemptive analgesic effect in the plantar incisional model of postsurgical pain in mice. Neurosci Lett. 2016;614:33–8.
31. Stache S, Close JD, Mehallo C, Fayock K. Nonprescription pain medication use in collegiate athletes: a comparison of samples. Phys Sportsmed. 2014;42(2):19–26.
32. Rudgard WE, Hirsch CA, Cox AR. Amateur endurance athletes’ use of non-steroidal anti-inflammatory drugs: a cross-sectional survey. Int J Pharm Pract. 2019;27(1):105–7.
33. Nieman DC, Henson DA, Dumke CL, et al. Ibuprofen use, endotoxemia, inflammation, and plasma cytokines during ultramarathon competition. Brain Behav Immun. 2006;20(6):578–84.
34. Klueglich M, Ring A, Scheuerer S, et al. Ibuprofen extrudate, a novel, rapidly dissolving ibuprofen formulation: relative bioavailability compared to ibuprofen lysinate and regular ibuprofen, and food effect on all formulations. J Clin Pharmacol. 2005;45(9):1055–61.
35. Yildiz HM, Speciner L, Ozdemir C, Cohen DE, Carrier RL. Food-associated stimuli enhance barrier properties of gastrointestinal mucus. Biomaterials. 2015;54:1–8.
36. Lima FD, Stamm DN, Della Pace ID, et al. Ibuprofen intake increases exercise time to exhaustion: a possible role for preventing exercise-induced fatigue. Scand J Med Sci Sports. 2016;26(10):1160–70.
37. Machida M, Takemasa T. Ibuprofen administration during endurance training cancels running-distance-dependent adaptations of skeletal muscle in mice. J Physiol Pharmacol. 2010;61(5):559–63.
38. Morgan PT, Vanhatalo A, Bowtell JL, Jones AM, Bailey SJ. Acute ibuprofen ingestion does not attenuate fatigue during maximal intermittent knee extensor or all-out cycling exercise. Appl Physiol Nutr Metab. 2019;44(2):208–15.
39. Da Silva E, Pinto RS, Cadore EL, Kruel LF. Nonsteroidal anti-inflammatory drug use and endurance during running in male long-distance runners. J Athl Train. 2015;50(3):295–302.
40. Fabricio de Souza R, Gama de Matos D, Pires Ferreira AR, et al. Effect of ibuprofen on muscle, hematological and renal function, hydric balance, pain, and performance during intense long-distance running. J Strength Cond Res. 2020;34(7):2076–83.
41. Morgan PT, Bowtell JL, Vanhatalo A, Jones AM, Bailey SJ. Acute acetaminophen ingestion improves performance and muscle activation during maximal intermittent knee extensor exercise. Eur J Appl Physiol. 2018;118(3):595–605.
42. Foster J, Taylor L, Chrismas BC, Watkins SL, Mauger AR. The influence of acetaminophen on repeated sprint cycling performance. Eur J Appl Physiol. 2014;114(1):41–8.
43. Mauger AR, Taylor L, Harding C, Wright B, Foster J, Castle PC. Acute acetaminophen (paracetamol) ingestion improves time to exhaustion during exercise in the heat. Exp Physiol. 2014;99(1):164–71.
44. Quinn CM, Duran RM, Audet GN, Charkoudian N, Leon LR. Cardiovascular and thermoregulatory biomarkers of heat stroke severity in a conscious rat model. J Appl Physiol (1985). 2014;117(9):971–8.
45. Lin Y-F, Wang J-Y, Chou T-C, Lin S-H. Vasoactive mediators and renal haemodynamics in exertional heat stroke complicated by acute renal failure. QJM. 2003;96(3):193–201.
46. Goppelt-Struebe M, Wolter D, Resch K. Glucocorticoids inhibit prostaglandin synthesis not only at the level of phospholipase A2 but also at the level of cyclo-oxygenase/PGE isomerase. Br J Pharmacol. 1989;98(4):1287–95.
47. Tamura M, Deb S, Sebastian S, Okamura K, Bulun SE. Estrogen up-regulates cyclooxygenase-2 via estrogen receptor in human uterine microvascular endothelial cells. Fertil Steril. 2004;81(5):1351–6.
48. Gregory MS, Duffner LA, Hahn EL, Tai HH, Faunce DE, Kovacs EJ. Differential production of prostaglandin E(2) in male and female mice subjected to thermal injury contributes to the gender difference in immune function: possible role for 15-hydroxyprostaglandin dehydrogenase. Cell Immunol. 2000;205(2):94–102.
49. Filaretova L. Glucocorticoids are gastroprotective under physiologic conditions. Ther Adv Chronic Dis. 2011;2(5):333–42.
50. Miyoshi H, VanDussen KL, Malvin NP, et al. Prostaglandin E2 promotes intestinal repair through an adaptive cellular response of the epithelium. EMBO J. 2017;36(1):5–24.


Supplemental Digital Content

Copyright © 2020 by the American College of Sports Medicine